© 2018

Composing Fisher Kernels from Deep Neural Models

A Practitioner's Approach


  • Presents a step-by-step approach to deriving a kernel from any probabilistic model belonging to the family of deep networks

  • Demonstrates the use of feature compression and selection techniques for reducing the dimensionality of Fisher vectors

  • Reviews efficient algorithms for large-scale image retrieval and classification systems, including concrete examples on different datasets

  • Provides programming solutions to help machine learning practitioners develop scalable solutions with novel ideas


Part of the SpringerBriefs in Computer Science book series (BRIEFSCOMPUTER)

Table of contents

  1. Front Matter
    Pages i-xiii
  2. Tayyaba Azim, Sarah Ahmed
    Pages 9-17
  3. Tayyaba Azim, Sarah Ahmed
    Pages 33-46

About this book


This book shows machine learning enthusiasts and practitioners how to get the best of both worlds by deriving Fisher kernels from deep learning models. In addition, the book shares insight on how to store and retrieve large-dimensional Fisher vectors using feature selection and compression techniques. Feature selection and feature compression are two of the most popular off-the-shelf methods for reducing data’s high-dimensional memory footprint and thus making it suitable for large-scale visual retrieval and classification. Kernel methods long remained the de facto standard for solving large-scale object classification tasks using low-level features, until the revival of deep models in 2006. Later, they made a comeback with improved Fisher vectors in 2010. However, their supremacy was always challenged by various versions of deep models, now considered to be the state of the art for solving various machine learning and computer vision tasks. Although the two research paradigms differ significantly, the excellent performance of Fisher kernels on the Image Net large-scale object classification dataset has caught the attention of numerous kernel practitioners, and many have drawn parallels between the two frameworks for improving the empirical performance on benchmark classification tasks. Exploring concrete examples on different data sets, the book compares the computational and statistical aspects of different dimensionality reduction approaches and identifies metrics to show which approach is superior to the other for Fisher vector encodings. It also provides references to some of the most useful resources that could provide practitioners and machine learning enthusiasts a quick start for learning and implementing a variety of deep learning models and kernel functions.


Deep Models Fisher Vectors Large Scale Information Retrieval Feature Compression Techniques Feature Selection Techniques

Authors and affiliations

  1. 1.Center of Excellence in ITInstitute of Management SciencesPeshawarPakistan
  2. 2.Institute of Management SciencesPeshawarPakistan

About the authors

Dr. Tayyaba Azim is an Assistant Professor at the Center for Information Technology, Institute of Management Sciences, Peshawar, Pakistan. 

Sarah Ahmed is a current research student enrolled in Masters of Computer Science program at Institute of Management Sciences Peshawar, Pakistan. 

She has received her  Bachelor’s Degree in Computer Science from Edwardes College, Peshawar,Pakistan. Her areas of interest include: Machine Learning, Computer Vision and Data-Science. Currently, her research work is centered around the feature compression and selection approaches for Fisher vectors derived from deep neural models. Her research paper: "Compression techniques for Deep Fisher Vectors" was awarded  the best paper in the area of applications at ICPRAM conference 2017. 

Bibliographic information

Industry Sectors
Chemical Manufacturing
IT & Software
Consumer Packaged Goods
Materials & Steel
Finance, Business & Banking